2 results listed
In the past two decades, human action
recognition has been among the most challenging tasks in
the field of computer vision. Recently, extracting accurate
and cost-efficient skeleton information became available
thanks to the cutting edge deep learning algorithms and
low-cost depth sensors. In this paper, we propose a novel
framework to recognize human actions using 3D skeleton
information. The main components of the framework are
pose representation and encoding. Assuming that human
skeleton can be represented by spatiotemporal poses, we
define a pose descriptor consists of three elements. The first
element contains the normalized coordinates of the raw
skeleton joints information. The second element contains
the temporal displacement information relative to a
predefined temporal offset and the third element keeps the
displacement information pertinent to the previous
timestamp in the temporal resolution. The final descriptor
of the skeleton sequences is the concatenation of frame-wis e
descriptors. To avoid the problems regarding high
dimensionality, PCA is applied on the descriptors. The
resulted descriptors are encoded with Fisher Vector (FV)
representation before they get trained with an Extreme
Learning Machine (ELM).
The performance of the proposed framework is evaluated
by three public benchmark datasets. The proposed method
achieved competitive results compared to the other methods
in the literature.
International Conference on Advanced Technologies, Computer Engineering and Science
ICATCES
Saeid Agahian
F.NEGIN
Cemal Kose
EEG signals are commonly used data sources in BCI
applications. For this reason, recent studies to analyze the EEG
signals in the most accurate way are increasing rapidly. When
features are extracted from EEG signals, the use of methods
sensitive to local variations is of great importance for correct
classification of the signals. In this study, 1D-local binary pattern
(LBP) method which is sensitive to local changes was applied to
motor imager/movement EEG signals and the obtained features
were classified with the k-NN and SVM classifiers. Accordingly, in
the case of using the k-NN method, the lowest 99.98%, and highest
100% classification accuracy was obtained.
International Conference on Advanced Technologies, Computer Engineering and Science
ICATCES
Funda Kutlu Onay
Cemal Kose